TCR interactions with self peptide-MHC complexes in the thymus control the key developmental steps of positive selection, lineage commitment and negative selection that are essential for the generation of a functional CD8 T cell repertoire. The underlying hypothesis of this grant is that to successfully pass through positive and negative selection, most thymocytes must interact sequentially with distinct sets of peptides that are presented on different thymic cellular elements. We postulate that specialized proteasomes (thymoproteasomes) in cTECs make peptides that are optimized for positive selection. We further hypothesize that to generate a broad and non-autoreactive CD8 T cell repertoire, it is generally necessary to switch the peptides used for negative selection in mTECs and DCs to ones that are different than used for positive selection (the proteasome/peptide switching hypothesis). To test these hypotheses, we have generated a novel mouse model that genetically lacks all 4 immuno- and thymo-proteasome subunits and therefore only expresses constitutive proteasomes in all cells. We propose to use this model together with state of the art genetic models, culture systems, and mass spectrometry to elucidate the underlying mechanisms of thymic selection.